UK: Nearly one in five Prevent referrals had autism, new Home Office figures reveal
UK: Nearly one in five Prevent referrals had autism, new Home Office figures reveal Submitted by Areeb Ullah on Thu, 08/06/2026 - 17:09 Critics question Prevent's reliance on subjective assessment after Home Office data reveals sharp rise in referrals involving autistic people Figures showed that nearly 10,000 people were referred to Prevent in England and Wales, the highest annual number of…
New Home Office figures reveal that nearly one in five individuals referred to the UK's Prevent counter-extremism program had autism, raising concerns over the scheme's impact on neurodivergent people. Between October 2024 and September 2025, 9,957 people were referred to Prevent in England and Wales, the highest annual total since records began in 2015.
Of these, 1,833 had autism recorded, with 1,199 confirmed cases and 634 suspected. Autism was the most commonly recorded neurodiverse condition, with 36 percent of all referred individuals having at least one mental health or neurodiversity condition. Critics argue that the increasing referrals of autistic individuals reflect longstanding issues with the government's counter-extremism program, which has been accused of bias and discrimination.
The Home Office data also showed a 20 percent rise in referrals related to far-right extremism and that Asian and Black people continue to be overrepresented in the program. The program's reliance on subjective judgement has been criticized, as public sector workers are legally obligated to refer anyone suspected of terrorism, potentially leading to biased decisions based on personal beliefs.
The Home Office emphasized that the data does not prove a causal relationship between neurodiversity and radicalization, but advocates warn that the program risks reinforcing stereotypes about autistic people and could have lasting consequences for those referred, even if no further action is taken.
Written by urgent.news from Middle East Eye's reporting — not their text. Machine-written — it may contain errors, so check the original before relying on it.